Path planning method and system for self-driving of autonomous system

1. A path planning method for self-driving of autonomous system, which is executed by an agent, the method comprising following steps of: S1: acquiring a path optimization function of the agent based on the driving space of the agent, a speed of the agent, and a current position of the agent; S2: converting, based on fixed-point theorems, the path optimization function of the agent into an equivalent fixed-point equation; S3: acquiring a complete simplex sequence based on the fixed-point equation; and S4: determining, based on the complete simplex sequence, an initial population size and an initial position of particles for particle swarm optimization to obtain the best path planning of the agent, and conducting the self-driving of the autonomous system according to the best path planning; wherein the step S2 is specifically: based on { min ⁢ ⁢ y = f ⁡ ( X ) st . ⁢ X ∈ { X | g i ⁡ ( X ) ≤ 0 , i = 1 , 2 , … ⁢ , m } , constructing a fixed-point equation F(X)=X−f(X), wherein, according to the fixed-point theorems, f(X*)=0 if X* is a solution to the fixed-point equation, so that a minimum value of the path optimization function y=f(X) at the point X* is obtained; where: f(X) is the path optimization function of the agent; X is an n-dimensional optimization variable; and g i (X) is m constraint functions in a function feasible region; according to the fixed-point theorems, introducing an approximate fixed point to replace a precise fixed point: let any ε>0, considering X as an approximate fixed point of the f(X) if |X−f′(X)|<ε, where|X−f′(X)| represents the modulus of a vector, specifically: S301: dividing a search space of the fixed-point equation, specifically: in an n-dimensional Euclidean space R n , dividing a search space of the fixed-point equation into uniform polyhedrons by a n-family straight line x i =mh i (1=1,2, . . . ,n), where m is precision control; S302: processing the divided search space by a simplicial algorithm to obtain simplexes, wherein: for the Euclidean space R n , N={1,2, . . . ,n}, π is the permutation of N, and n basis vectors u 1 , . . . ,u n of R n , which are n columns in an identity matrix of order n, satisfy the following condition: u=u 1 + . . . +u n =(1, . . . ,1); let K 1 0 be a set of integer points in R n , if y 0 ∈ K 1 0 , a n-dimensional simplex <y 0 ,y 1 , . . . ,y n > is denoted by k 1 (y 0 ,π), where y i =y i−1 +u π(i) , i∈N, and a set of k 1 (y 0 ,π) is denoted by K 1 ; and S303: labeling the simplexes to output a complete simplex sequence, specifically: labeling the simplexes by integer labeling or vector labeling, obtaining a complete simplex sequence satisfying labeling requirements according to a logistic discrimination, and using a value range of the complete simplex sequence as an updated search space; wherein the step S4 is specifically: in an n-dimensional solution space, if the position of each particle is a complete simplex sequence x i =(x i1 ,x i2 , . . . ,x in ), the number of complete simplex sequences represents the population size, the flying speed is v i =(v i1 ,v i2 , . . . ,v in ), the best position of individual particles is denoted by p best =(p i1 ,p i2 , . . . ,p in ), the best global position is g best =(p g1 ,p g2 , . . . ,p gn ), for the current particle, adjusting the current position x id and the current speed v id of this particle according to the following formulae: v id ⁡ ( t + 1 ) = ω ⁢ v id ⁡ ( t ) + c 1 ⁡ ( p id - x id ) + c 2 ⁡ ( p gd - x id ) x id ⁡ ( t + 1 ) = x id ⁡ ( t ) + v id ⁡ ( t + 1 ) { v id = v max , v id > v max v id = - v max , v id < - v max where: p id is the current best position of individual particles; p gd is the current best global position; v max represents the maximum flying speed of particles; −v max represents the minimum flying speed of particles; ω is an inertia weight; and c 1 and c 2 are acceleration constants.

2. The path planning method for self-driving of autonomous system according to claim 1 , wherein the value range of the complete simplex sequence is used as a value range of the flying speed.

3. The path planning method for self-driving of autonomous system according to claim 1 , wherein the inertia weight is set in the following way: an inertia decreasing weight is used, the value ω of which gradually decreases in iterations, specifically: ω = ω max - n · ( ω max - ω min ) n max , where: ω max =0.9; and ω min =0.4.

4. A path planning system for self-driving of autonomous system, the agent comprising: at least one storage unit; and at least one processing unit; wherein the at last one storage unit stores at least one instruction that, when loaded and executed by the at least one processing unit, implements the following steps: S1: acquiring a path optimization function of the agent based on the driving space of the agent, a speed of the agent, and a current position of the agent; S2: converting, based on fixed-point theorems, the path optimization function of the agent into an equivalent fixed-point equation; S3: acquiring a complete simplex sequence based on the fixed-point equation; and S4: determining, based on the complete simplex sequence, an initial population size and an initial position of particles for particle swarm optimization to obtain the best path planning of the agent, and conducting the self-driving of the autonomous system according to the best path planning; wherein the step S2 is specifically: based on { min ⁢ ⁢ y = f ⁡ ( X ) st . ⁢ X ∈ { X | g i ⁡ ( X ) ≤ 0 , i = 1 , 2 , … ⁢ , m } , constructing a fixed-point equation F(X)=X−f(X), wherein, according to the fixed-point theorems, f(X*)=0 if X* is a solution to the fixed-point equation, so that a minimum value of the path optimization function y=f(X) at the point X* is obtained; where: f(X) is the path optimization function of the agent; X is an n-dimensional optimization variable; and g i (X) is m constraint functions in a function feasible region; wherein the step S3 is specifically: according to the fixed-point theorems, introducing an approximate fixed point to replace a precise fixed point: let any ε>0, considering X as an approximate fixed point of f(X) if |X−f′(X)|<ε, where |X−f′(X)| represents the modulus of a vector, specifically: S301: dividing a search space of the fixed-point equation, specifically: in an n-dimensional Euclidean space R n , dividing a search space of the fixed-point equation into uniform polyhedrons by a n-family straight line x i =mh i (i=1,2, . . . ,n), where m is precision control; S302: processing the divided search space by a simplicial algorithm to obtain simplexes, wherein: for the Euclidean space R n , N={1,2, . . . ,n}, π is the permutation of N, and n basis vectors u 1 , . . . ,u n of R n , which are n columns in an identity matrix of order n, satisfy the following condition: u=u 1 + . . . +u n =(1, . . . ,1); let K 1 0 be a set of integer points in R n , if y 0 ∈K 1 0 , a n-dimensional simplex <y 0 ,y 1 , . . . ,y n > is denoted by k 1 (y 0 ,π), where y i =y i-1 +u π(i) , i∈N, and a set of k 1 (y 0 ,π) is denoted by K 1 ; and S303: labeling the simplexes to output a complete simplex sequence, specifically: labeling the simplexes by integer labeling or vector labeling, obtaining a complete simplex sequence satisfying labeling requirements according to logistic discrimination, and using a value range of the complete simplex sequence as an updated search space; wherein the step S4 is specifically: in an n-dimensional solution space, if the position of each particle is a complete simplex sequence x i =(x i1 ,x i2 , . . . ,x in ), the number of complete simplex sequences represents the population size, the flying speed is v i =(v i1 ,v i2 , . . . ,v in ), the best position of individual particles is denoted by p best =(p i1 , i2 , . . . ,p in ), the best global position is g best =(p g1 p g2 , . . . ,p gn ), for the current particle, the current position x id and the current speed v id of this particle is adjusted according to the following formulae: v id ⁡ ( t + 1 ) = ω ⁢ v id ⁡ ( t ) + c 1 ⁡ ( p id - x id ) + c 2 ⁡ ( p gd - x id ) x id ⁡ ( t + 1 ) = x id ⁡ ( t ) + v id ⁡ ( t + 1 ) { v id = v max , v id > v max v id = - v max , v id < - v max where: p id is the current best position of individual particles; p gd is the current best global position; v max represents the maximum flying speed of particles; −v max represents the minimum flying speed of particles; ω is an inertia weight; and c 1 and c 2 are acceleration constants.